Document worth reading: “An Essay on Optimization Mystery of Deep Learning”

Despite the massive empirical success of deep finding out, theoretical understanding of neural networks finding out course of stays to be lacking. This is the reason, why some of its choices seem ‘mysterious’. We emphasize two mysteries of deep finding out: generalization thriller, and optimization thriller. In this essay we analysis and draw connections between a quantity of chosen works regarding the latter. An Essay on Optimization Mystery of Deep Learning